Abstract
In this issue of Cephalalgia, Dr Jakob Moller Hansen and his colleagues report a series of very interesting experiments in which they successfully induced migraine-like attacks in patients with migraine with aura via infusion of calcitonin gene related peptide (CGRP). This work echoes similar experiments carried out by the Copenhagen group using the nitric oxide donor, glycerol trinitrate (GTN) (1), and builds on previous findings from the group that infusion of CGRP in patients suffering from migraine without aura induced a delayed headache with features of migraine (2).
CGRP is a 37 amino acid neuropeptide (3) that is expressed in neurons arising from the trigeminal ganglion (4), which has been associated with cephalic pain for almost two decades. CGRP is a potent vasodilator (5), and is one of several neuropeptides known to be released from activated perivascular dural c-fibres (6,7). It may mediate the vasodilatory component of neurogenic inflammation and modulate trigeminovascular pain transmission, although, in at least one animal model, CGRP did not activate or sensitise meningeal nociceptive neurons (8). The link between CGRP and migraine in humans has been a little more tenuous. There has been conflicting data as to whether CGRP levels in external jugular blood are elevated during migraine attacks in humans (9–11). However, the induction of migraine-like attacks by CGRP infusion reported by Dr Hansen and others would certainly seem to strengthen the link between CGRP and human migraine. The reported efficacy of the new CGRP antagonist Telcagepant in acute migraine further suggests the potential importance of CGRP in migraine (12).
The series of CGRP infusion experiments reported by Dr Hansen and colleagues was carried out in patients who reported that they experienced attacks of migraine with aura exclusively. These infusions differ from an earlier series carried out in migraine without aura (2), in that they infused both migraine patients and non-migraine control subjects with CGRP, and compared the relative frequency of headache induction. Whereas the earlier series was carried out in migraine without aura patients, and utilised a blinded cross-over design in which migraine sufferers (but not non-migraine control subjects) received infusions of either placebo or CGRP. After the infusion, the occurrence of delayed headache was compared between CGRP and placebo arms. These experiments both provide pertinent data but not identical data. The Hansen group experiments provide information about the relative vulnerability of migraine with aura subjects to migraine attack induction by CGRP. In contrast, the Lassen series carried out exclusively in migraine without aura and non-migraine subjects gives information about the relative likelihood of headache induction by CGRP. In addition, another series of CGRP infusions was carried out in familial hemiplegic migraine (FHM) subjects and normal controls using a design similar to the experiments in migraine with aura (13). Interestingly, CGRP-induced delayed attacks of migraine-like headache in patients with typical aura and without aura but not in familial hemiplegic migraine.
From these data, Dr Hansen and colleagues postulate that the mechanism of headache induction in migraine with typical aura and migraine without aura are likely to be similar, but that the mechanism in familial hemiplegic migraine may be distinct. Their findings in this series would tend to support this hypothesis. It would be interesting to determine, after the administration of CGRP under identical experimental conditions, if patients suffering from migraine without aura share a similar relative susceptibility and time course for migraine attack as those with migraine with aura. The hypothesis that the pathophysiological mechanisms of the common forms of migraine (with and without aura) differ in some fundamental way from that of familial hemiplegic migraine would also seem to be indirectly supported by the prolonged nature of the neurological symptoms that occur during hemiplegic migraine. The hemiplegia may last for days and persist long after resolution of the headache. However, if we are to assume there is such a difference, how do we account for the co-occurrence in most patients with FHM of clinically typical auras? If in existence, the possible mechanistic differences between FHM and the common forms of migraine will be a fascinating new area for research.
The induction in this series of CGRP infusions of migraine aura in four of 14 patients is clearly different from the findings of previous similar GTN experiments (1) and raises the question of how CGRP induction may differ from migraine induction by nitric oxide donation.
The experiments of Hansen and colleagues provide important new data in our quest to understand migraine better and are quite timely as a CGRP antagonist may soon be added to our antimigraine armamentarium (12). As with all good science, the findings from these experiments raise many new and interesting questions.